and their never-ending supports. This study was a one-way design where the independent variable was animation interactivity. In addition to a control group (Static Group) provided with only static materials, there were three groups with different levels of animation interactivity: 1) Animation with simple interactivity (Simple Animation Group), 2) animation with input vii manipulation (Input Group), and 3) animation with practice and feedback (Practice Group). A sample of 123 college students participated in the study and was randomly assigned into groups. They gathered in the computer lab to work with the animation program and then took online surveys and tests for evaluation. Students were expected to learn Principles of Hypothesis Testing (concepts of type I error, type II error and pvalue). The data collected in this study included 1) student learning attitudes, 2) achievement and confidence pre-test scores, 3) achievement and confidence post-test scores, and 4) program perception. Also, student manipulation of the animation program was recorded as Web log data. The data were analyzed by using multivariate analysis (MANOVA), univariate analysis (ANOVA), regression analysis, regression tree analysis and case analysis.The findings were as follows: 1) Animation interactivity impacted students' improvement in understanding (p=.006) and lower-level applying (p=.042), 2) animation interactivity did not impact student confidence and program perception, 3) the regression analysis indicated that student prior knowledge and interest were the most important predictors on student achievement post-test scores instead of program manipulation, and 4) the regression tree showed that there were interactions among student interest, prior knowledge, and program manipulation on the achievement post-test scores. The case analysis showed that not all students manipulated the interactive animation program as expected due to a lack of motivation and cognitive skills, and this could decrease the effect of the interactive animation. This study hoped to broaden theories on interactive learning and serve as a reference for future statistics curriculum designers and textbook publishers.viii He asked students to first clarify the definition of the type I and type II error, and then look for clues in the question for computation. The type I and type II error problems are difficult for most novice learners. But with the concrete examples of "bags," "boxes,"
List of Tablesor "envelopes," statistics was no longer as difficult for me. This particular teaching method is valuable for use in all introductory-level classes.
3I believe this struggle with learning statistics is not limited to myself, but is common to many other students. As a result, I created an online animated program based on my instructor's teaching method in order to help learners who also struggle with learning type I error, type II error and the p-value. This online animation was created to popularize this particular teaching method. To benefit learners with different ...